Minelab GP minimum sample time self-adjust?

[bklein]

How does a GP series adjust to various coil selections - I mean it seems the minimum sample time somehow gets recalibrated with each coil attached or even each power application. The PI's we build have the minimum sample time manually set and it may have to be tweaked by us with a different coil. Since some of the designs here are starting to use microcontrollers, what would be the algorithm to have our DIY detector optimize this setting?

[Qiaozhi]

How does a GP series adjust to various coil selections - I mean it seems the minimum sample time somehow gets recalibrated with each coil attached or even each power application. The PI's we build have the minimum sample time manually set and it may have to be tweaked by us with a different coil. Since some of the designs here are starting to use microcontrollers, what would be the algorithm to have our DIY detector optimize this setting?

I suspect the detector is not actually doing any sort of clever coil calibration, but is instead interrogating the coil to find out which type is attached, and then setting the minimum delay accordingly. The "interrogation" may be as simple as detecting which pins are connected together in the coil plug.

[sido]

Just a thought. Maybe when powered up the I-R-C are calculated from the given coil connected, then the detector adjusts the correct minimum sample delay for optimum calibration as such.

Varying sizes of coils will generally have the Inductance spot on, but the Resistance and Capacitance may differ slightly, therefore can be calculated in uC and minimum timings adjusted accordingly.

Cheers Sid

[joop]

I suspect the detector is not actually doing any sort of clever coil calibration, but is instead interrogating the coil to find out which type is attached, and then setting the minimum delay accordingly. The "interrogation" may be as simple as detecting which pins are connected together in the coil plug.

Maybe just an RFID tag, like they are using with toner cartridges...

[bklein]

The gp coils I have taken apart did not have any identifiers like pin wiring or RFID. The detector does some form of optimization on power up. There is a button on the Extreme to force this on demand. The fact that diy'rs are successful making their own coils supports a self optimization routine.

[Qiaozhi]

The gp coils I have taken apart did not have any identifiers like pin wiring or RFID. The detector does some form of optimization on power up. There is a button on the Extreme to force this on demand. The fact that diy'rs are successful making their own coils supports a self optimization routine.

I've just looked at the user manual for the GP Extreme, and I cannot see any mention of automatic calibration of the coil for minimum sample delay.

Are you referring the "Tune" button?
If so, then this control is used to automatically tune out the external EMI by selecting the most stable range of frequencies, and is something completely different.

[bklein]

Yes but I am not saying it does - just suggesting it does. Even what you suggest tune is doing doesn't make much sense on a pi detector...

[Woody.au]

All GP SD GPX Transmit timings are fixed and only change on the GPX series by using the Enhance / Fine Gold switch.
They do not sense the coil, only in regard to no coil connected or over current by using a very low resistance coil or shorted coil.

[bklein]

The TX timings would be fixed but I am wondering if the RX sample timings are optimised. I find it hard to believe that RX timing of first sample is fixed irregardless of coil dynamics. It isn't even fixed in a surf PI - it is a manual adjustment. So really, the timing is just set by firmware to a default and never changed? If so, I would think that could be tweaked for improved detection with faster coil designs.

[Woody.au]

The settings do not self adjust to optimise the receive window, any nasty ground that gets into the receive window is as it is. The detector just gets noisy. The problem being that the ratio between everything TX RX is set in firmware.... So by adding different tx lengths changes all aspects of rx and dead time by linear contraction or extension. Many of the parameters need to be operator adjustable, If you are working nasty ground increase the dead time, working Pipe Clay bring it in. Big coil? Longer TX , Small Coil bring it in. So many combinations and full performance will never be realised with the fixed relationships of the unit.

[sido]

The settings do not self adjust to optimise the receive window, any nasty ground that gets into the receive window is as it is. The detector just gets noisy. The problem being that the ratio between everything TX RX is set in firmware.... So by adding different tx lengths changes all aspects of rx and dead time by linear contraction or extension. Many of the parameters need to be operator adjustable, If you are working nasty ground increase the dead time, working Pipe Clay bring it in. Big coil? Longer TX , Small Coil bring it in. So many combinations and full performance will never be realised with the fixed relationships of the unit.

So does any given (uP) have any control of the coil parameters in use or is it a stand alone user adjustable format as you say with ones timings??

Interesting, considering that by nowdays of the newer technological implementations and advances in innovation, that some sort of self adjusting Micro-controlled platform would and could perform the required timing, sampling pulses and GB elimination to achieve a crucial and optimum adjustments for peak performance..

Quote from Woody
"The problem being that the ratio between everything TX RX is set in firmware.... So by adding different tx lengths changes all aspects of rx and dead time by linear contraction or extension"

Can you please elaborate on what you are trying to say above?? Apart from the obvious.. thank you

Cheers sid